Mammalian somatic cells tightly regulate the genes they express, yet have the capacity to be reprogrammed to express a different spectrum of genes. This plasticity of gene expression is readily apparent upon fusion with other somatic cells or following nuclear transfer. For example, it is possible to induce a specialized adult cell type, such as a hepatocyte, to express previously silent genes typical of another specialized cell type such as muscle, by fusing it to a multinucleated myotube. Thus, it is becoming increasingly clear that the maintenance of the differentiated state is ongoing and largely controlled by dynamic mechanisms, requiring continuous regulation. Specialization of cells for function in particular tissues is not a fixed and irreversible property, but one that can be altered if nuclei are exposed to the appropriate cocktail of intracellular factors. These changes in cell phenotype cross ectodermal, mesodermal, and endodermal lineages. Of key importance is the balance of regulators present in a cell at any given time, as this dictates the expression patterns of genes characteristic of particular tissues. An understanding of the intracellular regulatory network will enhance our ability to enlist this potential for change and deliberately direct cells towards a specialized state of interest.

The two major approaches to studying the mechanisms underlying the reprogramming of mammalian nuclei are cell fusion and nuclear transfer. The Blau laboratory uses both to examine the effects of diverse intracellular regulators on the patterns of genes that nuclei express. To examine nuclear reprogramming, nuclei are exposed to different combinations of factors, either by micromanipulation (nuclear transfer) or using polyethylene glycol (cell fusion). Heterokaryons, stable multinucleate fusion products of mouse and human cells provide an ideal system for study, as they do not exhibit cell or nuclear division, or chromosome loss. As a result of fusion, nuclei derived from a diversity of cell types, for example hematopoietic stem cells, are exposed to a panoply of muscle or neural intracellular factors. The critical roles of certain regulators in chromatin remodeling and gene expression, are then studied by a combination of gene silencing (siRNA) and overexpression using state-of-the-art viral vectors for delivery of genes to non-dividing cells. These studies are unveiling the importance of specific repressors and activators that act independent of cell division, directly at the level of DNA or indirectly as modifiers or remodellers of chromatin. Especially instructive are the profound differences among cell types in the role of particular chromatin remodeling steps.

Cell fusion to form heterokaryons occurs spontaneously in mice and in humans. By tracking hematopoietic stem cell fate after a bone marrow transplant, we and others recently found that derivatives of HSCs fuse specifically with Purkinje neurons in the brain, skeletal muscle fibers, and hepatocytes. The frequency of this event increases substantially in the presence of selective pressure and following tissue damage. In the Purkinje neurons of the brains of women who had received a bone marrow transplant from a male donor for the treatment of leukemia, Y-chromosomes indicative of fusion were found post-autopsy. These neurons are highly complex, with a full dentritic tree, synapses to millions of other neurons, and a critical role in balance and movement. Since, Purkinje neurons are not made anew after birth, one testable hypothesis is that cell fusion could serve as a means of cell rescue by nuclear replacement. We are studying mice with Purkinje defects to address this possibility.

Taken together, our studies are providing evidence that not only oocytes, but also postnatal somatic cells, have the molecular machinery required for reprogramming nuclei and that cell fusion may have practical applications for deriving desired cell types for use in cell based therapies.

Bone marrow contribution to Purkinje neurons in the cerebellum

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

In the cerebellum of a mouse post-bone marrow transplanted, individual donor-derived GFP+ Purkinje neurons are evident in the Purkinje cell layer. The dendrites from these cells extend into the cell sparse molecular layer, while their axon projects through the granular cell layer. Laser scanning confocal image of this cell show its many synaptic spines and single output axon.

Bone marrow contribution to myofibers in muscles

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Contribution of bone marrow derived cells (BMDC) expressing GFP (green) to skeletal muscles (blue). GFP-expressing fibers were identified as myofibers by their clear sarcomeric distribution and large volumes which were visualized in three-dimensional image stacks captured with a laser scanning confocal microscope. Small GFP+ blood cells are evident (white arrows). No GFP myofibers nor blood cells were ever observed in tissue sections of skeletal muscles from mice successfully transplanted with non-GFP bone marrow.

Immunohistochemistry of skeletal myofibers

expressing bone marrow derived GFP

 

 

 

 

 

 

 

 

 

 

 

 

 

Thin optical sections captured using laser scanning confocal microscopy to conclusively demonstrate that GFP-expressing myofibers have characteristic staining patterns on the cell surface for laminin (red). No GFP+ skeletal myofibers but most GFP+ cells (i.e., blood cells) in skeletal muscle expressed either CD45, a marker of hematopoietic cells, or F4/80 antigen, a marker of macrophages.

Bone marrow contribution to hepatocytes in liver

 

 

 

 

 

 

 

In situ hybridization and immunofluorescent staining of hepatocyte-specific human alpha-1 antitrypsin (hAAT). Binucleate hepatocyte obtained from the liver of a female FVB mouse post-bone marrow transplanted with hAAT+ transgenic male donor. Fluorescent in situ hybridization with Y-chromosome specific probe (red) confirmed its derivation from male donor and hAAT staining (green) demonstrate the participation of the bone marrow-derived nucleus to the hepatocyte program.

Bone marrow contribution to Muscle

 

 

 

 

 

 

 

 

 

 

 

In these images, which were captured with a confocal microscope, the optical thickness of each image is 2-3 ƒÊm. All myofibers were assessed for nuclear location using a three dimensional analysis of a stack of thin optical sections and nuclei were required to be contained within the three dimensional volume of the myofiber in order to be counted. Thus, nuclei which appear centrally located in GFP+ (arrows) and non-GFP expressing (arrowheads) myofibers are unquestionably myonuclei contained within the myofibers and not overlying blood cell nuclei.